第六章供应链中的需求预测

1-1Copyright©2013PearsonEducation.Copyright©2013PearsonEducation.Copyright©2013PearsonEducation.1-1Copyright©2013PearsonEducation.1-1Copyright©2013PearsonEducation.7-1Copyright©2013PearsonEducation.第七章DemandForecastinginaSupplyChain7-2Copyright©2013PearsonEducation.LearningObjectives1.Understandtheroleofforecastingforbothanenterpriseandasupplychain.2.Identifythecomponentsofademandforecast.3.Forecastdemandinasupplychaingivenhistoricaldemanddatausingtime-seriesmethodologies.4.Analyzedemandforecaststoestimateforecasterror.7-3Copyright©2013PearsonEducation.RoleofForecastinginaSupplyChain•Thebasisforallplanningdecisionsinasupplychain•Usedforbothpushandpullprocesses–Productionscheduling,inventory,aggregateplanning–Salesforceallocation,promotions,newproductionintroduction–Plant/equipmentinvestment,budgetaryplanning–Workforceplanning,hiring,layoffs•Allofthesedecisionsareinterrelated7-4Copyright©2013PearsonEducation.CharacteristicsofForecasts1.Forecastsarealwaysinaccurateandshouldthusincludeboththeexpectedvalueoftheforecastandameasureofforecasterror2.Long-termforecastsareusuallylessaccuratethanshort-termforecasts3.Aggregateforecastsareusuallymoreaccuratethandisaggregateforecasts4.Ingeneral,thefartherupthesupplychainacompanyis,thegreateristhedistortionofinformationitreceives7-5Copyright©2013PearsonEducation.ComponentsandMethods•Companiesmustidentifythefactorsthatinfluencefuturedemandandthenascertaintherelationshipbetweenthesefactorsandfuturedemand–Pastdemand–Leadtimeofproductreplenishment–Plannedadvertisingormarketingefforts–Plannedpricediscounts–Stateoftheeconomy–Actionsthatcompetitorshavetaken7-6Copyright©2013PearsonEducation.ComponentsandMethods1.Qualitative–Primarilysubjective–Relyonjudgment2.TimeSeries–Usehistoricaldemandonly–Bestwithstabledemand3.Causal–Relationshipbetweendemandandsomeotherfactor4.Simulation–Imitateconsumerchoicesthatgiverisetodemand7-7Copyright©2013PearsonEducation.ComponentsofanObservationObserveddemand(O)=systematiccomponent(S)+randomcomponent(R)•Systematiccomponent–expectedvalueofdemand−Level(currentdeseasonalizeddemand)−Trend(growthordeclineindemand)−Seasonality(predictableseasonalfluctuation)•Randomcomponent–partofforecastthatdeviatesfromsystematiccomponent•Forecasterror–differencebetweenforecastandactualdemand7-8Copyright©2013PearsonEducation.BasicApproach1.Understandtheobjectiveofforecasting.2.Integratedemandplanningandforecastingthroughoutthesupplychain.3.Identifythemajorfactorsthatinfluencethedemandforecast.4.Forecastattheappropriatelevelofaggregation.5.Establishperformanceanderrormeasuresfortheforecast.7-9Copyright©2013PearsonEducation.Time-SeriesForecastingMethods•Threewaystocalculatethesystematiccomponent–MultiplicativeS=levelxtrendxseasonalfactor–AdditiveS=level+trend+seasonalfactor–MixedS=(level+trend)xseasonalfactor7-10Copyright©2013PearsonEducation.StaticMethodsSystematiccomponent=(level+trend)´seasonalfactorFt+l=[L+(t+l)T]St+lwhereL=estimateoflevelatt=0T=estimateoftrendSt=estimateofseasonalfactorforPeriodtDt=actualdemandobservedinPeriodtFt=forecastofdemandforPeriodt7-11Copyright©2013PearsonEducation.TahoeSaltYearQuarterPeriod,tDemand,Dt1218,00013213,00014323,00021434,00022510,00023618,00024723,00031838,00032912,000331013,000341132,000411241,000Table7-17-12Copyright©2013PearsonEducation.TahoeSaltFigure7-17-13Copyright©2013PearsonEducation.EstimateLevelandTrendPeriodicityp=4,t=3Dt=Dt–(p/2)+Dt+(p/2)+2Dii=t+1–(p/2)t–1+(p/2)åéëêêùûúú/(2p)forpevenDi/pforpoddi=t–[(p–1)/2]t+[(p–1)/2]åìíïïîïïDt=Dt–(p/2)+Dt+(p/2)+2Dii=t+1–(p/2)t–1+(p/2)åéëêêùûúú/(2p)=D1+D5+2Dii=24å/87-14Copyright©2013PearsonEducation.TahoeSaltFigure7-27-15Copyright©2013PearsonEducation.TahoeSaltFigure7-3AlinearrelationshipexistsbetweenthedeseasonalizeddemandandtimebasedonthechangeindemandovertimeDt=L+Tt7-16Copyright©2013PearsonEducation.EstimatingSeasonalFactorstttDDSFigure7-47-17Copyright©2013PearsonEducation.EstimatingSeasonalFactorsSi=Sjp+1j=0r–1årS1=(S1+S5+S9)/3=(0.42+0.47+0.52)/3=0.47S2=(S2+S6+S10)/3=(0.67+0.83+0.55)/3=0.68S3=(S3+S7+S11)/3=(1.15+1.04+1.32)/3=1.17S4=(S4+S8+S12)/3=(1.66+1.68+1.66)/3=1.67F13=(L+13T)S13=(18,439+13´524)0.47=11,868F14=(L+14T)S14=(18,439+14´524)0.68=17,527F15=(L+15T)S15=(18,439+15´524)1.17=30,770F16=(L+16T)S16=(18,439+16´524)1.67=44,7947-18Copyright©2013PearsonEducation.AdaptiveForecasting•Theestimatesoflevel,trend,andseasonalityareadjustedaftereachdemandobservation•Estimatesincorporateallnewdatathatareobserved7-19Copyright©2013PearsonEducation.AdaptiveForecastingFt+1=(Lt+lTt)St+1whereLt=estimateoflevelattheendofPeriodtTt=estimateoftrendattheendofPeriodtSt=estimateofseasonalfactorforPeriodtFt=forecastofdemandforPeriodt(madePeriodt–1orearlier)Dt=actualdemandobservedinPeriodtEt=Ft–Dt=forecasterrorinPeriodt7-20Copyright©2013PearsonEducation.StepsinAdaptiveForecasting•Initialize–Computeinitialestimatesoflevel(L0),trend(T0),andseasonalfactors(S1,…,Sp)•Forecast–Forecastdemandforperiodt+1•Estimateerror–ComputeerrorEt+1=Ft+1–Dt+1•Modifyestimates–Modifytheestimatesoflevel(Lt+1),trend(Tt+1),andsea